> On 03/08/2011 07:50 PM, Bogdan Szczurek wrote:
> >> On Tue, Mar 8, 2011 at 3:12 PM, Bogdan
> >> Szczurek<thebod...@gmail.com> wrote:
> >>> I also have high hopes for GEGL, but I'd rather have it use some
> >>> more abstract color model for that. I know it's not so
> >>> simple—maybe even undoable–that way, but I do like the idea of
> >>> color model that has complete coverage of visible spectrum.
> >>
> >> Using a color model with full coverage of the visual spectrum
> >> would be an extension along the lines of RGB and the responses of
> >> the human visual system/physics, entirely additive.
> >
> > Not entirely along the lines I'm afraid. First of all it strongly
> > depends which RGB we're talking about. Even if we take scRGB into
> > account there's still some considerable parts of visible spectrum
> > that can not be described by scRGB's triad. I know scRGB is
> > tempting for it's quite broad and seems easy to implement in RGB
> > dominated world, but I've got a premonition that using device
> > agnostic color space would pay off more. But again–I don't know
> > that :).
> If all you want is a color space that can encode all visible colors, 
> i.e. the entire CIEXYZ color space, RGB is fine. Transforming from 
> CIEXYZ to RGB (and vice versa) is a simple matrix multiplication,
> where the matrix depends on the primaries and white point chosen.
> It's just that sometimes the RGB components will be negative and
> sometimes greater than 1.0, but each color that can be perceived by a
> human can be encoded in such a boundless RGB color space.

Yes, but why use RGB at all if one can use e.g. XYZ from the start?
So "wide" RGB would also require greater than 8 bit depths to work
satisfactorily (HSV, HSL or Lab do quite nicely even with 8 bits per
component). I think one consolation is possible backward compatibility
with some other RGB spaces, but isn't BC a b**ch? ;) Besides there's a
trouble of defining such “new” RGB workspace: what white point should
be choosen and what primaries (all have to be defined in some absolute
color coordinates BTW)? Whatever space would be choosen we wouldn't
escape color conversions in color managed workflow, so while I'm not
RGB enemy I fail to see the reason to stick to it especially since
there are “wider” color spaces that are more intuitive to work with.

> If you want a color model that doesn't lose the information about the 
> spectral power distribution of the stimulus, then RGB won't do, but
> from your reply above it doesn't sound like that is what you meant.

No, I didn't, but still I think RGB “could do” since it's able to
describe absolute color.

Well… I don't know if my longing is good or not, or if that way is
better—I'm just thinking aloud :).

My best!

Attachment: signature.asc
Description: PGP signature

Gimp-developer mailing list

Reply via email to